{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/dcfnet-discriminant-correlation-filters","title":"DCFNet: Discriminant Correlation Filters Network for Visual Tracking","arxiv_id":"1704.04057","date":"2017-04-13","proceeding":null,"authors":["Qiang Wang","Jin Gao","Junliang Xing","Mengdan Zhang","Weiming Hu"],"abstract":"Discriminant Correlation Filters (DCF) based methods now become a kind of\ndominant approach to online object tracking. The features used in these\nmethods, however, are either based on hand-crafted features like HoGs, or\nconvolutional features trained independently from other tasks like image\nclassification. In this work, we present an end-to-end lightweight network\narchitecture, namely DCFNet, to learn the convolutional features and perform\nthe correlation tracking process simultaneously. Specifically, we treat DCF as\na special correlation filter layer added in a Siamese network, and carefully\nderive the backpropagation through it by defining the network output as the\nprobability heatmap of object location. Since the derivation is still carried\nout in Fourier frequency domain, the efficiency property of DCF is preserved.\nThis enables our tracker to run at more than 60 FPS during test time, while\nachieving a significant accuracy gain compared with KCF using HoGs. Extensive\nevaluations on OTB-2013, OTB-2015, and VOT2015 benchmarks demonstrate that the\nproposed DCFNet tracker is competitive with several state-of-the-art trackers,\nwhile being more compact and much faster.","url_abs":"http://arxiv.org/abs/1704.04057v1","url_pdf":"http://arxiv.org/pdf/1704.04057v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"dcfnet-discriminant-correlation-filters","repo_url":"https://github.com/foolwood/DCFNet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"dcfnet-discriminant-correlation-filters","repo_url":"https://github.com/HaHuangChan/CACFNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"dcfnet-discriminant-correlation-filters","repo_url":"https://github.com/QiHuangChen/CACFNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"dcfnet-discriminant-correlation-filters","repo_url":"https://github.com/linzhi123/DCFNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"dcfnet-discriminant-correlation-filters","repo_url":"https://github.com/linzhi123/DCFNet-pytouch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"object-tracking","task_name":"Object Tracking"},{"task_slug":"visual-tracking","task_name":"Visual Tracking"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[{"method_slug":"heatmap","method_name":"Heatmap"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1704.04057","atlas_url":"https://app.syntology.ai/?focus=1704.04057","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}